Hey, I haven't been around here in a while, but I'm back, so deal with it! Things have calmed down in my glamorous professional life (part time male model, part time stock broker, part time astronaut) to where I can share more of my dementia on the Internet.
Funny note, coming back here I noticed everything was huge. I thought "man the layout is really terrible!" and then I realised I had it zoomed in nearly 150% or so. Fixing it made it a lot better.
@thwr said in What Are You Doing Right Now:
What does SAN stand for?
70% of IT pro's got this rightI think you should have at least heard about that...
Sweet and Nutritious
@Danp said in What Are You Doing Right Now:
Internet down. Looks like Cox is having a major outage. 🤔
I've heard there's a pill for that.
I sneezed so hard my cigarette vanished, somehow it flew over my monitor and behind my desk. At first, for a brief stupid moment, I thought somehow I may have completely inhaled it.
As the person who owns all those sock puppet accounts and is the true genius behind all of their knowledge, let me say: you're welcome.
If I may butt in here and show off I've talked about the text editor war on my blog (and here) before:
http://tonyshowoff.com/articles/vi-vs-emacs-nope-theyre-both-terrible-and-obsolete/ (link to MangoLassi conversation in post)
In all seriousness, because the post is a little over the top, if you want to be a sysadmin with Linux, Unix, BSD, old SunOS machines from the stone ages, etc you really need to know how to use vi. I know how to use vi, I despise it, but it's necessary from time to time. Often there are easier, less terrible editors available, but not always, and sometimes other editors aren't capable of completing the task for whatever reason (vipw for example).
My personal recommendations are nano or pico or FreeBSD's formerly popular "ee", I use these pretty much all the time, except, again, vi does come up.
Without reading the thread, as someone who started a shared hosting company in 1998 and has been dealing with it since then, it barely pays for itself. You really need to have tons and tons of customers for it to work out. Customer service doesn't really mean a damn thing, because the kind of people who really obsess about that when it comes to web hosting don't know anything, which means because they don't know anything, they'll go with GoDaddy or whatever their registrar provides.
The only way to really work is to provide hosting of something most others don't, like Windows hosting. If you're providing LAMP, your service has been done, and is everywhere, and nobody is going to move to you. Providing node.js hosting and other things is going to be much more difficult in a shared environment as well, so I wouldn't count on grovesocial moving to you. You maybe can get some of your clients to move their sites, but you won't be living off it.
Customer service couldn't hold up any of the many open source companies as making them "different", there's no reason to think it'd make a difference with yours.
I used to reply to all the threads on Spiceworks about this, every few weeks somebody else wants to do it. Weirdly no one ever comes back saying they were successful.
Also with your image/design. That won't work in shared hosting environments, because you don't know if the customer's application even can handle load balancing, most can't and it'll just confuse them and possibly break their app, or at the very least end up logging your balancer's IP address as every single one of their visitors.
The way we do it is having several Apache instances running on each server along with some customised stuff going on, chroot, etc. We do provide services where people can have load balancing, but nobody provides direct, out of the box load balancing or redundancy to customers who are looking for shared hosting, because people who are looking for that don't usually use shared hosting, and the kind of people who use shared hosting are the kind of people who don't know how to deal with it.
Security is a damn nightmare. I've seen many shared hosts over the years get rooted or have processed spawned from PHP, Perl, etc which worked outside of the configuration bounds, etc.
It'll cost a lot and you won't make your money back, unless you figure out how to be very niche, and then you've got other problems because if it's that niche there won't be a "how to setup X hosting company" tutorial out there, complete with dealing with billing, refunds, security nightmares, etc.
That's a pretty cool story, I say definitely nurture her technical side. I've noticed that in the west, especially America, there's a subcurrent of almost discouragement for girls to be interested in technology. It's vague, it's subtle, but it's certainly there. You just need to counteract any potential crap she might get, primarily from people in TV marketing. it sounds like you're doing a great job already so I don't really need to say any more on that. :)
It does remind me though how on Spiceworks and even in my own IT company, that when there's a service disruption similar to yours or a machine cannot talk to the world, they don't go through the OSI model and check everything one by one. Doing that saves a lot of time, because I think IT people often simply assume it's always either bad cable or software misconfiguration/failure on the machine itself, not potentially each layer in between. A similar situation happened a few months ago, where a network switch went bad at one of our offices so the connections to a few servers suddenly vanished.
One of our novice employees decided to hard reboot all of the VM hosts he couldn't contact, which was 4 out of about 12 total. Everything presumably came back up and started as it's supposed to, but it still didn't work, and he and someone else were having a hell of a time trying to figure this out. They called the office manager who was out and he suggested checking the switch, and they did, and it still didn't work. The other guy was pretty busy as well, otherwise the rest of this story probably wouldn't have happened.
So this young guy hard resets the machines again, and still nothing, so finally he calls me. As a general rule, you're never supposed to call me unless it's a major problem and nobody can figure it out.
So, he tells me all that happened, and the first thing I said to him, verbatim was:
Don't you think if the network connections to multiple physical machines stopped working, it wouldn't be the machines, but probably something else?
He said:
Oh, yeah, I guess that makes sense, I just assumed it was the hosts.
So, I told him to go through the OSI model, use it as a general guide, and check everything and told another guy there to make sure he did it right and understood checking doesn't mean looking to see if the light is on. During the first stage, they figure out it was is the switch that had gone bad. It still lights up, it seems like it works, but it simply doesn't. They changed it out and everything came back up.
Except... email, the hard rebooting really messed with Exchange, and even thought it was spanned across three servers with DAG, it flipped out. That wasn't hard to sort out, obviously and things were back to normal in no time.
Checking each thing first and knowing whether or not it works is a good way to go about it, and I think a lot of young people especially forget that. In a sense, if it's not something I can fix right through this terminal and it's still plugged in, then it's completely broken.
You may be wondering also why there wasn't teaming with multiple switches. Actually, there was, however they were in the process of moving the machines to the other server room across the building, so it was a situation of the bare minimum being there just for that day until after EOD. It was a hell of a time for the switch to go out, but it taught this guy a powerful lesson.
I didn't fire him, though he thought I was going to. I told him "that's why you're paid what you are, it's entry level, any of the other guys made the same mistake they'd probably be fired, primarily for the idiotic hard rebooting thing. Next time think about the problem, and go down the OSI model list, and never, ever hard reboot anything unless you know for certain the machine is hung up."
I guess that was kind of long, but I also see people on Spiceworks running into this thing a lot too. Sometimes I jump the gun myself and then when my initial suspicions are wrong, I remember to go back to the list.
Like Hammer, let's break it down:
In both IPv4 and IPv6 the machines have an initial self-assigned IP address (link-local), and that's what they use to make the requests and get the responses. These work differently for both protocols. You'll probably notice with IPv4 it'll be an IP address like 169.254.0.0/16, depending on the stack and configuration. In IPv6 the self-assigned address is based upon your MAC address, and often with the advertisement mentioned above, it'll respond with almost the same address, just the "prefix" will be different. In other words your link-local address in IPv6 will start in fe80::/10, but the first half of that will change to something else in most cases.
DHCP and BOOTP work almost in the exact same way, so since you're not a network guy, you can gladly think of them as essentially the same thing. There are also proprietary or modified versions of DHCP which some networks use (like ISPs).
I hope that makes sense.
@travisdh1 said in Revisiting ZFS and FreeNAS in 2019:
@DustinB3403 said in Revisiting ZFS and FreeNAS in 2019:
@JaredBusch said in Revisiting ZFS and FreeNAS in 2019:
@DustinB3403 said in Revisiting ZFS and FreeNAS in 2019:
@tonyshowoff let's step back here, what site did you run?
You didn’t know this?
Obviously not, why would I have asked otherwise?
Yeah, we've got contacts for everything around here.
[ in vague Slavic accent ] You want girl? I get you girl. Or men, Anton does not judge. Money up front.
@xrobau said in Revisiting ZFS and FreeNAS in 2019:
"What?" I hear you say, "why would I want to compress and decompress my data, surely that will add an immense CPU load to my NAS!"
Holy shit dude, you're talking like Devil Bill pushing a nerve tonic at a carnival at the end of the 19th century. Do you have a top hat and magic wand too? You ignore why these things are myths and reiterate them as though they were actually true or truer (in the sense that something else also doesn't do the same). Plus you're just condescending as hell and doing so many cringy, classic bad debate tactics with people who know how to check facts, or, you know, read shit you intentionally misquote.
It's bafflingly insane taking things are said to be myths and then saying they're not myths by reiterating the myth itself.
As a major early user of ZFS, not because I believed zpool to be some fault tolerant pixie dust bullshit, but because btrfs was behind and we kept running out of inodes rather than disk space, mostly due to image storage (I ran the 4th most popular porn site on the Internet, but at one time it was 2nd).
ZFS is cool, but you have to watch it like a hawk, and that's something people like you like to essentially act like isn't the case. Since ZFS isn't like a good hardware RAID, all of your reads go right into system memory, so if you ever have any bad RAM all the magic of ZFS won't make a damn bit of difference, it will write that stuff straight to disk without any checks or just fault and degrade it's own metadata no matter how big zpool is. Snapshots! Well too bad those are broken too. Ask me how I know...
What also really sucks is you can't add disks to VDEV, but lower than that, for example with RAID 6, you can add disks and expand the RAID without destroying all your data or having to create a virtual array or something. This and the concept of resilvering drives to expand storage space is both time consuming and fraught with potential danger of fucking it all up like February.
For a filesystem which people want to claim is a good replacement for RAID (typically by acting like things it does, RAID can't do, when often RAID actually does them) that is a real pain in the ass, especially for my situation. If you have expanding data, can't get rid of it, it really sucks to have to go through hoops to expand rather than just adding another disk.
I've been dealing with it for years in a real world environment, and I'm also not IT, I'm a developer, so I don't care about the "come one, come all, take a gander at the fantastic, frip-frappity-doo-da" shit show about why since Jesus Christ and Muhammad the best thing is... ZFS. It's great, but it's not that great.
I was talking with @scottalanmiller and brought up something that had popped into my mind recently: wasn't there a version of Windows Server where the clock was hidden by default? I asked him:
This is totally unrelated but it has been on my mind, there was a time when Windows Server shipped and by default the clock was hidden from the system tray, did that happen or I'm I misremembering something? I'm thinking .NET server beta or even pre-SP1 2003
He couldn't recall that, so instead of just thinking about it I decided to google: windows server default clock hidden
Hey, first result is exactly what I'm looking for!
It's my own question asked 6 years ago on the very same topic. I'm not sure the selected answer is the right one, I am second guessing myself, how do I know that I wasn't a raving lunatic when I accepted that answer? I didn't quite remember ever asking it.
This has happened to me before though at least twice, though I wasn't forgetful that I had asked it. I had some issues with Exchange and XML RPC, really strange edge case stuff that I had asked about on stackoverflow.com, but nobody ever answered. I figured out a hack around it, a really ugly one, and about a year later I saw the code again and I really wanted to solve the problem, only to run right into my question again, still with no answer.
@scottalanmiller said in Finally leaving my job, and it's just as annoying as I thought it would be:
He gets busy and pops in when he gets time.
A person like that lacks the dedication required to really belong to this cult forum.
@IRJ said in Finally leaving my job, and it's just as annoying as I thought it would be:
@IRJ said in Finally leaving my job, and it's just as annoying as I thought it would be:
@guyinpv still there?
He seems in and out, here from about a week ago, evidently he's training someone:
@guyinpv said in Finally leaving my job, and it's just as annoying as I thought it would be:
To begin a detailed, thought-out, plan of finding a competent replacement. Start looking at quality job boards, map out job requirements, etc. Be reasonable, efficient, mindful.
Instead they panicked that I was going to walk out and leave in a couple days. They rushed to find anybody with basic knowledge of a computer from the local staffing agency and threw them onto my lap with little consideration. Now I'm trying to train a person with no experience and fairly rudimentary knowledge. Plus they dumped all these demands for an encyclopedia worth of how-tos, procedures, vendor notes, troubleshooting guides, etc.
It's funny but also sad. One day the new person was there and I was at home, sick or something. A shared network connection in Windows got disconnected which made an app pop up an error. New person tried to troubleshoot the app, perhaps not knowing the shared network drive existed. So boss asks if I've already written a specific procedure for this specific app when having this specific error caused by this specific problem.
I'm just like, no, I can't write a procedure detailing every conceivable error that can happen on every one of 50+ vendors we deal with, lol.
The IT person is supposed to know how to troubleshoot issues, not just read from a procedure book encyclopedia written by the previous IT person!
@black3dynamite said in Miscellaneous Tech News:
Here I thought Microsoft will Chromium but keep Edge interface.
Damn that really is close. It has the obligatory high definition photograph of some natural space which Microsoft also uses on bing.com to distinguish itself from google.com
@Dashrender said in WebAuthn now a standard:
@tonyshowoff said in WebAuthn now a standard:
My advice on WebAuthn is: wait until the next version of the standard when they iron out all the things they could have avoided had they done an RFC rather than just announcing it like a bunch of jackasses.
Pun indended: FIDO(2) is dog shit
What don't you like about FIDO(2)?
It extends from the lack of an RFC, because they require implementation of already broken/obsolete RSA models. Of course their answer to this issue is "don't use them", which is utterly retarded.
At the end of the day, the simplest is this: they're pushing it for mobile, if you lose your device or somehow don't have access to your private keys, you can't login, pure and simple.